EViews.com

This thread is about dccgarch11 add-in.

The add-in allows you to build and estimate Dynamic Conditional Correlation models, which are the more flexible and parameterized class of Multivariate GARCH-family. It is written/designed with primarily educational purposes in mind and therefore some limitations are imposed to ease the estimation and maintain the user-friendliness of the GUI as well:

i) Second or higher order specifications are not allowed.
ii) Estimation is carried out in two-step.
iii) At most 5 series are allowed.

Please read the documentation for further instructions.

Edit: EViews had made a significant overhaul to its optimization engine starting from version 9, which resulted in messy outputs of the add-in thereafter. Please update your copy with the newest version of the add-in.

very useful Add-in . thanks a lot.

is the documentation available?

trubador wrote:This thread is about dccgarch11 add-in.

The add-in allows you to build and estimate Dynamic Conditional Correlation models, which are the more flexible and parameterized class of Multivariate GARCH-family. It is written/designed with primarily educational purposes in mind and therefore some limitations are imposed to ease the estimation and maintain the user-friendliness of the GUI as well:

i) Second or higher order specifications are not allowed.
ii) Estimation is carried out in two-step.
iii) At most 5 series are allowed.

Please read the documentation for further instructions.

Assuming you have already installed the add-in, the documentation is available from the Manage Add-ins menu.

Hi, I am new to eviews. I would need to simultaneously estimate a VEC and a bivariate DCC-GARCH, I have try to use the add-in for dcc-garch but I cannot get it to work, also I would need the VEC output in line with the dcc output. would I have to right the program myself or is there is any other way?

Dear trubador
Does this need Eviews 8 or higher to go with?

Thanks.

Yes, you need EViews version 8 or higher.

trubador wrote:Yes, you need EViews version 8 or higher.

Dear Trabadur, thanks for this add-in, is there any specific reason that only 5 variables are allowed?

Have you read the documentation that comes with the add-in?

[quote="trubador"]Have you read the documentation that comes with the add-in?[/quo

Unfortunately I cant open it, it keep sending me error!

PDF document should be stored here: C:\Users\...\Documents\EViews Addins\dccgarch11\dccgarch11.pdf

trubador wrote:PDF document should be stored here: C:\Users\...\Documents\EViews Addins\dccgarch11\dccgarch11.pdf

Thank you very much

Before estimating a DCC-GARCH(1,1) model, time series have to be filtered to assure zero expected (mean) value of the time series. Usually, a bivariate Vector Autoregressive (VAR) model used to initially remove potential linear structure, then the residuals of the VAR model are used as inputs for the DCCGARCH model.
Is this procedure involved in the add-in or do I need to do this separately and give residuals as input to the add-in?

After running the add-in, how do I get graphical plot of DCC GARCH conditional correlation ( say on y-axis) with time (say on x-axis) between two series?

If your questions are regarding the general use of the add-in, just post them under the relevant thread in this forum. Please do not send private e-mail messages (especially not to my office account).

akash27 wrote:Before estimating a DCC-GARCH(1,1) model, time series have to be filtered to assure zero expected (mean) value of the time series. Usually, a bivariate Vector Autoregressive (VAR) model used to initially remove potential linear structure, then the residuals of the VAR model are used as inputs for the DCCGARCH model.Is this procedure involved in the add-in or do I need to do this separately and give residuals as input to the add-in?

The add-in only allows univariate AR-X models for mean equations. VAR model needs to be specified at the outset.

akash27 wrote:After running the add-in, how do I get graphical plot of DCC GARCH conditional correlation ( say on y-axis) with time (say on x-axis) between two series?

If you select "Save dynamic correlations" option, then you can plot either rho_??_?? for the series of interest or rhos?? for all correlations post estimation. For instance, for a two-variable model it would be: